1-s2.0-S0037073806002077-main

7/28/2019 1-s2.0-S0037073806002077-main

1/16

Processes, facies and architecture of fluvial distributary

system deposits

G.J. Nichols a,b,, J.A. Fisher a

aDepartment of Geology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, U.K.

b The University Centre in Svalbard, P.O. Box 156, Longyearbyen N-9171, Norway

Abstract

There is evidence from the stratigraphic record of examples of fluvial deposits that were the products of deposition from river

systems which had decreasing discharge down-flow and transitions from proximal, channelised to distal, unconfined flow. These

deposits form fan-shaped bodies several tens of kilometres in radius, and their stratigraphic architecture is aggradational, with no

evidence of deep incision driven by base-level fall. The fluvial systems that generated these deposits formed under conditions for

which there is no complete analogue today: an endorheic basin with a relatively arid climate adjacent to an uplifted area with higher

precipitation. A conceptual model for fluvial systems of this type has therefore been built on the basis of outcrop examples and a

consideration of the controls on sedimentation. Proximal areas are characterised by amalgamated coarse, pebbly and sandy channel

deposits with little preservation of overbank facies. Channel dimensions are generally smaller in the medial areas, but sizes are

variable: deposits are of braided, meandering and simple channels which show varying degrees of lateral migration. The channel-

fills may be mud or sand, with overbank flow processes playing an important role in filling channels abandoned on the floodplainafter avulsion. The proportion of overbank deposits increases distally with sheets of sand deposited as lateral and terminal splays by

unconfined flow. Interconnection of sandstone bodies is poor in the distal areas because channel-fill bodies are sparse, small and

are not deeply incised. The radial pattern of the sediment body forms by the repeated avulsion of channels: active channels build up

lobes on the alluvial plain and rivers switch position to follow courses on lower lying areas. The term fluvial distributary system is

here used to describe a river system which has a downstream decrease in discharge and has a distal zone which is characterised

either by terminal splays on to a dry alluvial plain or a lake delta during periods of lake highstand.

2006 Elsevier B.V. All rights reserved.

Keywords: Fluvial distributary system; Terminal fans; Fluvial fans; Endorheic basins; Fluvial stratigraphic architecture

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

2. Characteristics of a fluvial distributary system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

2.1. Dimensions of the sedimentary unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

2.2. Proximal facies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

2.3. Medial facies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Sedimentary Geology 195 (2007) 7590

www.elsevier.com/locate/sedgeo

Corresponding author. Department of Geology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, U.K.

E-mail address: [email protected](G.J. Nichols).

0037-0738/$ - see front matter 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.sedgeo.2006.07.004
mailto:[email protected]://dx.doi.org/10.1016/j.sedgeo.2006.07.004http://dx.doi.org/10.1016/j.sedgeo.2006.07.004mailto:[email protected]

7/28/2019 1-s2.0-S0037073806002077-main

2/16

2.4. Distal facies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

2.5. Distributary pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

3. Fluvial channel and overbank processes in a distributary system . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3.1. Trends in fluvial channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3.2. Discharge variations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3.3. Floodplain deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3.4. Floodplain channel-fills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3.5. Distal zone sheets: terminal splays and floodouts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

3.6. Formation of a fan of fluvial deposits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

3.7. Bifurcation of channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

3.8. Proximal to distal extent of channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

4. Fluvial distributary systems and lakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

5. Conditions for the formation of a fluvial distributary system: tectonic and climatic setting . . . . . . . . . . . . . . 85

6. Fluvial distributary systems, terminal fans, fluvial fans, megafans and lake deltas . . . . . . . . . . . . . . . . . . . 86

6.1. Alluvial fans, fluvial fans, megafans, humid fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

6.2. Subaerial and losimean fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

6.3. Terminal fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

6.4. Lake deltas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

6.5. Fluvial distributary systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 877. The stratigraphic architecture of fluvial distributary system deposits . . . . . . . . . . . . . . . . . . . . . . . . . . 87

8. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

1. Introduction

Friend (1978) pointed out that some ancient river

systems seem to show features that distinguish themfrom many modern-day river systems. He recognised

three distinctive characteristics that could be recognised

in ancient fluvial stratigraphic units: (1) a downstream

decrease in river depth,(2) an absence of alluvial incision

and (3) a convex-upwards, lobate topography of the river

systems. He suggested that these features indicated

deposition by a distributive river system which formed a

terminal fan. Subsequent work in one of the areas of

Friend's original studies, the Ebro Basin, established

more details of the sedimentology of two fluvial

distributary systems of Miocene age (Hirst and Nichols,1986; Nichols, 1987; Friend, 1989; Hirst, 1991). The

concept of terminal fans was also expanded by Kelly and

Olsen (1993) with reference to some Devonian exam-

ples. In this paper we summarise the architectural

characteristics of fluvial systems of this type and

consider the tectonic, climatic and related base-level

controls on their formation and preservation. Examples

from the Miocene of the northern Ebro Basin are used to

illustrate the characteristics of these systems, which are

typically tens of kilometres in radius and are comprised

of fluvial channel and overbank deposits which vary in

relative abundance and character between proximal and

distal areas. Comparison is also made with other

subaerial fan deposits, including alluvial fans, fluvial

fans and megafans, and the usage of different terminol-

ogy considered. The term fluvial distributary system isused in preference to terminal fan for reasons which are

discussed in later sections.

2. Characteristics of a fluvial distributary system

Conceptual models for a fluvial distributary system are

shown in Figs. 1 and 2. These have been developed from

earlier models presented in Friend (1978), Nichols (1987,

1989), Kelly and Olsen (1993) and Stanistreet and

McCarthy (1993) with the addition of data from other

sources (e.g. Graham, 1983; MacCarthy, 1990; Sadler andKelly, 1993; Williams, 2000; Nichols, 2004, 2005; Fisher

et al., 2006-this volume) and new observations in the Ebro

Basin. The locations of the examples from the strati-

graphic record used in this review are shown in Fig. 3.

2.1. Dimensions of the sedimentary unit

The radial distance from the apex to the distal fringe

of the system is in the order of tens of kilometres: the

Luna and Huesca systems in the Miocene of the Ebro

Basin are respectively 40 and 60 km radius (Hirst and

Nichols, 1986), the systems in the Devonian Munster

76 G.J. Nichols, J.A. Fisher / Sedimentary Geology 195 (2007) 7590

7/28/2019 1-s2.0-S0037073806002077-main

3/16

Basin in southern Ireland are up to 120 km radius

(Williams, 2000) and the system in the Wood Bay

Formation of the Devonian of Spitsbergen is up to200 km long (Friend and Moody-Stuart, 1972). The

extent of a system would be determined by the size of

the basin and the balance between water supply and loss

due to evaporation/infiltration, so it would be possible to

have deposits covering a larger area than the examples

given. The thickness of the succession will be limited by

the accommodation in the basin below the spill point,

the level at which water flows out of the basin and it

ceases to be endorheic (Bohacs et al., 2000), and the

sediment supply. There is approximately 4000 m

thickness of fluvial strata in the northern part of theEbro Basin (Nichols, 1987) and the Munster Basin fill is

over 6000 m thick (Graham, 1983; MacCarthy, 1990;

Sadler and Kelly, 1993).

2.2. Proximal facies

The coarsest deposits occurring in the deepest

channels are found in the proximal zones of the systems.

These are sandy or conglomeratic facies showing clast

imbrication, cross-bedding and preservation of bar

structures indicating that they are the deposits of braided

streams close to margin of the basin (e.g. Graham, 1983;

Nichols, 1987; MacCarthy, 1990; Sadler and Kelly,

1993). The proximal channel-fills in the Luna System are

reported to be up to 7 m thick (Nichols, 1987) and in theMunster Basin the channel-fill successions are at least

10 m thick (MacCarthy, 1990). In the Devonian of East

Greenland, proximal zone facies in the Snehvide

Formation are sandy braidplain succession of medium-

grained, trough-cross bedded pebbly sandstones and fine

pebble conglomerates (Friend et al., 1983). In these and

other proximal zone examples, there are no associated

fine-grained overbank deposits preserved, and the

channel-fill facies are entirely amalgamated. This

indicates that the channels were mobile, laterally

migrating and reworking the adjacent floodplain depos-its or repeatedly avulsing to new positions on the

proximal floodplain area. The interconnectedness of the

coarse facies is therefore 100% in the proximal zone.

Stacks of sandstone and conglomerate can be recognized

in borehole core as the deposits of the proximal parts of a

fluvial distributary system, for example in the Devonian

strata in the Clair Basin, West of Shetland (Nichols,

2005). The extent of the proximal zone is limited to

between 5 and 10 km from the system apex in the Luna

System (Nichols, 1987) whilst in the Munster Basin

there are conglomeratic fluvial deposits over 40 km from

the basin margin (MacCarthy, 1990; Williams, 2000).

Fig. 1. Proximal to distal trends in channel and overbank processes across a fluvial distributary system: channel forms, floodplain deposits and

architectural elements. The proximal to distal extent of the system is likely to be tens of kilometres.

77G.J. Nichols, J.A. Fisher / Sedimentary Geology 195 (2007) 7590

7/28/2019 1-s2.0-S0037073806002077-main

4/16

2.3. Medial facies

The transition from the proximal to the medial zone

is marked by an increase in the proportion of overbank

facies preserved and a decrease in the clast size of

channel-fill facies. In the Munster Basin the channel

facies are commonly trough-cross bedded sandstones

and pebbly sandstones occurring in multi-storey com-

plexes tens of metres thick (Graham, 1983; Sadler and

Kelly, 1993). The bases of the channel complexes are

not deeply incised into the floodplain facies and areinterpreted as the deposits of laterally mobile bedload

streams (Sadler and Kelly, 1993). Medial zone deposits

in the Ebro Basin are rarely pebbly, and are also partly

interpreted as the deposits of laterally mobile streams.

However, Hirst (1991) has demonstrated that the width

of the channel-fill sandstone bodies decreases down-

stream: at 25 to 35 km from the apex of the Huesca

System 80 to 90% of the sandstone bodies have width:

thickness ratios N15:1 whereas in the more distal

exposures 50 to 60 km from the apex only 5 to 35%

have width:thickness ratios N15:1. These data indicate

that the channels tended to become more laterally stable

down-system, a trend also noted by MacCarthy (1990)

in the Munster Basin.

The percentage of deposits in-channel was docu-

mented from measurements made of over 250 channel

sandstone bodies across the Huesca System in the Ebro

Basin by Hirst (1991). In the medial area, at 25 km from

the calculated apex 62% of the deposits occupied

channels, and this decreased to values of 30 to 40%

between 40 and 45 km, and further decreased to between

10 and 15% at distances of 45 to 50 km from the apex.

Similar trends have been qualitatively established in theDevonian examples from Ireland (Graham, 1983;

MacCarthy, 1990; Sadler and Kelly, 1993), Spitsbergen

(Friend and Moody-Stuart, 1972) and East Greenland

(Friend et al., 1983).

The channel-fill bodies are enclosed within strata

which are overall finer-grained, consisting mainly of

mudrocks and thin sheet bodies of sandstone (Fig. 4a, b).

These overbank deposits may show evidence of

desiccation or soil formation. Calcareous pedogenic

features are noted from the purple siltstone floodplain

facies in the Munster Basin (Graham, 1983). The pale

brown mudrocks which dominate the overbank facies

Fig. 2. The development of a fan-shaped body of sediments by repeated avulsion of the river channel and the architectural characteristics of the

proximal, medial and distal zones of a fluvial distributary system. The area of deposition is likely to be tens of kilometres radius.


7/28/2019 1-s2.0-S0037073806002077-main

5/16

association in the Ebro Basin show relatively weakly

developed palaeosol profiles (Nichols, 1987). The

sandstone sheets are thin, typically only a few centi-

metres thick, and laterally extensive over tens of metres:

they have sharp bases and show some current ripple

cross-lamination. Sandstone sheets tend to occur in

packages separated by metres of mudrock.

2.4. Distal facies

The distal reaches form the most distinctive elements

of the fluvial distributary systems. They are charac-

terised by a very high proportion of floodplain facies,

with channel-fill deposits comprising only a few percent

of the strata. In the Huesca System, the percentage, by

volume, of deposits in-channel was shown to be 10%

or less at distances of over 50 km from the apex of the

system (Hirst, 1991). In the Luna System, Nichols(1987) reported that the thickness of the channel-fills are

on average less than in the medial zone, decreasing to an

average of just over 2 m. Distal facies in the Munster

Basin similarly display fewer and thinner channel-fills

than the proximal and medial zones (Graham, 1983;

Sadler and Kelly, 1993), and many of these channels are

shallow and rather poorly defined.

In the Ebro Basin artificial exposures created by road

and canal cuts allow the internal geometries and fills of the

channels to be examined in exceptional detail (see Fig. 4).

Channel scours wholly or partly filled with mudrock areseen to be relatively common: some are simple scour and

fill deposits, others show evidence of lateral migration of

the channel, with sand deposited on lateral accretion

surfaces on the inner bank of a meander bend and a final

fill with mud (Fig. 4d). Channel scours of a variety of sizes

which are wholly filled with sandstone also occur (Fig.4c)

and the processes of formation of these deposits are

considered in a later section.

A prominent feature of the distal zones is that much

more sandstone is present as thin sheet deposits than as

channel-fill facies. These sheets have sharp, sometimes

erosive base and this scouring at the base of the sheetssuggest local channelisation of flow (e.g. Fig. 4g;

Graham, 1983). Isolated sheets may occur, but mostly

they are found in packages of laterally extensive units

(Fisher et al., 2006-this volume). Between the sheet

sandstone beds, thick-bedded muddy and silty facies

show evidence of palaeosol development (Sadler and

Kelly, 1993; Fisher et al., 2006-this volume).

2.5. Distributary pattern

A characteristic of these fluvial systems is that thepattern of flow indicated by palaeocurrent data has a

radial, distributive form. This was first noted by Friend

(1978) and a subsequent statistical study of palaeocurrent

readings from 350 palaeochannels in the Luna System by

Jupp et al. (1987) demonstrated that an apex lying at the

basin margin could be defined. In other examples the

channel pattern may be elongate and less clearly radial,

(e.g. in the Munster Basin, Williams, 2000) but the

distributive character of the channels is a common feature.

The implication of these patterns is that each of the

systems was supplied by a river which entered the basin at

a point along the margin, and then spread out onto the

Fig. 3. Locations of examples of fluvial distributary systems from (a)

the Devonian (palaeogeographic reconstruction from Friend et al.,

2000) and (b) the Miocene of the Ebro Basin, Spain.


7/28/2019 1-s2.0-S0037073806002077-main

6/16

Fig. 4a-c. Examples of channel and overbank facies from the Luna and Huesca Systems, Miocene, Ebro Basin, Spain. (a) Sandstone bodies representing the

part of the Huesca System. This is in an area of the medial part of the system where channel-fill deposits represent between 30 and 40% of the succession

and 3 m thick in the medial Luna System. The fill of the lower channel is partly eroded by the later channel, which shows a complete fill of sandstone, interp

text). (c) A channel scour 2.5 m deep filled largely with mudstone: the scour cut into sandy channel deposits (right) and overbank facies (left).

7/28/2019 1-s2.0-S0037073806002077-main

7/16

Fig. 4d-g. Examples of channel and overbank facies from the Luna and Huesca Systems, Miocene, Ebro Basin, Spain. (d) Sandstone with lateral accretion s

2 m deep channel in the medial to distal parts of the Huesca System: the right side of the channel-fill is a mixture of mudstone and sandstone beds. (e) A sm

System deposits, filled with sand when the channel lays abandoned on the floodplain. (f) Mudstone and a sheet sandstone incised by small channel scour les

Huesca System. (g) Sheet sandstone bodies up to 0.5 m thick in the distal Luna System: some show erosive bases (centre), and most are composite, m

7/28/2019 1-s2.0-S0037073806002077-main

8/16

alluvial plain. More than one entry point for a distributary

system is indicated by Arenas et al. (2001) for the Luna

System, although it is possible that each was active at

different stages in the evolution of the system. In the

Munster Basin there may have been several coeval entry

points (Williams, 2000).

3. Fluvial channel and overbank processes in a

distributary system

The fluvial channel and overbank facies described

from the examples of fluvial distributary systems in the

stratigraphic record can be used to determine the de-

positional processes which are acting in different parts

of a system. These may be used to develop a conceptual

model for an active fluvial distributary system (Figs. 1

and 2).

3.1. Trends in fluvial channels

The apex of the system lies at a point along a basin

margin where drainage from the hinterland is funnelled into

the basin (Hirst and Nichols, 1986). A gradient change as

the river enters the basin promotes deposition of sediment.

River channels in the proximal areas are typically bedload-

dominated and may be pebbly or sandy. Cross-bedding and

the preservation of bar forms suggest deposition in braided

rivers (Collinson, 1996): discharge must be sufficient to

carry coarse gravel, and over long enough periods forbraided river bar forms to develop. The absence or limited

preservation of overbank facies indicates mobile rivers

repeatedly avulse or migrate to different positions near the

apex. Strong currents in the channel scour the banks and

inhibit the growth of vegetation which could stabilise the

channel margins. The area near the apex of the system

would therefore be expected to have the appearance of a

braidplain, somewhat similar to a gravelly glacial outwash

deposit (Boothroyd and Ashley, 1975; Boothroyd and

Nummedal, 1978).

Downstream the proportion of gravel in the riverchannel decreases and the bedload is mainly sand

deposited on mid-channel bars, with gravel limited to

basal lags. The depth of the channel decreases and it

may become less laterally mobile: floodplain areas are

less frequently occupied by channels, allowing vegeta-

tion to have a more pronounced stabilising effect on the

banks. The role of vegetation on bank stability would

not have been a factor in pre-Devonian times, and may

have had a limited effect in the Devonian (cf. Fraticelli

et al., 2004). A progressive reduction in discharge

downstream due to evaporation and infiltration of water

into the floodplain occurs along with a decrease in the

grain size of material carried and a decrease in the river

gradient. The channel dimensions become progressively

smaller, and there is a transition from bedload to mixed

load character. As the outer edge of the medial zone is

reached the river channel becomes sinuous, showing

evidence of meandering, straight or possibly anasto-mosing habit (Stanistreet and McCarthy, 1993).

3.2. Discharge variations

Fluvial systems may be subject to fluctuations of

discharge because of seasonal rainfall variations in the

hinterland catchment area. The high rates of water loss

through evaporation and soak-away in a fluvial distributary

systemmeanthat during periods of lower discharge theouter

parts of the system may not receive much or any water, even

if there is still some supply to the proximal zone. Evidenceof discharge fluctuations is therefore to be expected, espe-

cially in the medial and distal parts of the system. A channel

may experience periods of slack water during which mud

will be deposited in the shallow parts of the channel: this

may be preserved as drapes on bars and is particularly well

displayed on the lateral accretion surfaces of meandering

rivers (Fig. 4d). Total desiccation of a channel may occur,

with evidence of this preserved as mudcracks on mud layers

within a channel-fill succession (Nichols, 1987).

3.3. Floodplain deposition

In the proximal zone the floodplain consists of

abandoned channel deposits, but in the medial zone

overbank deposits are preserved. These are mostly mud

deposited out of suspension during flood events, but

there are also extensive sheets of sand deposited as

splays from the channel margins. The thickness and

extent of the sand sheets vary with the magnitude of the

flood, and the bases of some show signs of scouring into

the underlying sediment (Fisher et al., 2006-this

volume). The sands may show parallel lamination or

be current rippled, but generally they are structureless,possibly because of bioturbation or pedoturbation. Soils

develop on the floodplain, but do not develop into

mature profiles if rates of floodplain accumulation are

relatively high. Aeolian reworking of the alluvial plain

has not been recognised in the Ebro Basin, but has been

noted in some Devonian basins (Kelly and Olsen, 1993;

Richmond and Williams, 2000; Nichols, 2005).

3.4. Floodplain channel-fills

Sandy bedload may be preserved within a braided or

meandering river if the active channel migrates laterally or


7/28/2019 1-s2.0-S0037073806002077-main

9/16

the discharge decreases through time, resulting in a partial

fill of the channel with sand (Fig. 4d). In-channel depo-

sition may cause choking and trigger flooding and/or the

avulsion of a river to a new position. The processes of

avulsion leave a stretch of channel isolated from direct

supply of sediment and water from the active channel. If theisolated, inactive channel tract is a relatively short distance

from the active river, it may be filled with sand which fills

the depression until it is level with the floodplain surface.

Some simple sandstone channel-fills (Fig. 4e, f) are

probably entirely deposits of overbank deposition from

another channel, not the deposits of the confined flow

which created the scour. The wings which are preserved

on the margins of some sandstone bodies (Friend et al.,

1979) can sometimes be shown to be continuous with sand

which filled the upper parts of the channel-fill body: the

upper channel and the wing deposits are likely to havebeen formed after the channel was abandoned, and are the

products of overbank flow from another, active channel.

An abandoned channel that lies at a greater distance from

an active channel may be filled with mud deposited out of

suspension from flood events to form a clay plug (Fig.

4c). The frequency of occurrence of mud-filled channels

is likely to be underestimated in the stratigraphic record

because it may be difficult to recognise them within a

muddy overbank succession. A further possibility in the

distal segments of a distributary system is a fill of a

channel scour by lacustrine deposits if there is a lake

level rise and the lake margin floods onto the overbankarea. Therefore, although some of a channel body may

be sediment that was deposited by in-channel flow, the

final fill is more likely to be by unconfined flow after

abandonment.

3.5. Distal zone sheets: terminal splays and floodouts

The distal reaches of the fluvial distributary systems

are analogous to the floodout zones in Tooth (1999a,b).

In the ephemeral rivers of arid central Australia, river

channels commonly terminate in areas where thechannel capacity is reduced and floodwaters spread

across the alluvial surface forming terminal floodouts

(Tooth, 1999a,b): these features are characterised by a

transition from channelised to unchannelised flow. The

floodouts in central Australia occur on a wide range of

scales, from 1 km2 to 1000 km2, depositing sheets of

fine-grained sediment. Tooth (1999a) showed that their

occurrence is determined by the presence of barriers to

flow, such as the presence of a build up of aeolian sands,

or they may form where there is a downstream decrease

in discharge. Both mechanisms are possible in the distal

reaches of an aggrading fluvial system.

Features similar to terminal floodouts are also recog-

nised in the Neales River, which is on the western side of

the Lake Eyre Basin in central Australia.These are referred

to as terminal splays (Lang et al., 2004) and they are

sheets of sand deposited by decelerating flows on the

floodplain and on the dry lake bed. Their location iscontrolled by the incision of the river channel by base-level

fall driven by falling lake level, and the river flow becomes

unconfined at the edge of the dry lake bed (Lang et al.,

2004). These terminal splay deposits show horizontal

stratification, current ripple lamination and show subaque-

ous dune structures in places; the sheet sandstones of the

distal zones of the ancient distributary systems such as the

Munster Basin deposits (Sadler and Kelly, 1993) and Ebro

Basin deposits are typically structureless, horizontally

laminated or rarely ripple cross laminated.

A significant difference between eitherthe floodouts orthe terminal splays of central Australia and the sheets of

sediment formed in the distal zone of the ancient fluvial

distributary systems documented above, is the sediment

supply. In central Australia the sediment supply is very

low and the rivers are flowing very close to bedrock,

which is exposed in places. In contrast, the distal fluvial

distributary system deposits are documented from basins

which have thick accumulations of sediment, and the

preservation of both channel and overbank facies suggests

higher rates of sediment supply. The ancient floodouts/

terminal splays are therefore analogous in terms of

processes, but the architecture of the deposits may notbe comparable with the modern examples because of

differences in sediment supply and other variables, such

as flow magnitude.

3.6. Formation of a fan of fluvial deposits

The distributary system is built up as a series of

incremental units deposited by individual rivers which

flow from the apex to a point on the floodplain where their

flow becomes unconfined (Fig. 2). During the lifetime of

this river it will deposit sediment within the channel, onthe adjacent areas of the floodplain during periods of

overbank flow, and at the distal end of the channel as

terminal splays onto the alluvial plain. The river may

laterally migrate during this time, and local channel

abandonment may occur due to meander cut-off or local

readjustments of the position of the channel on the plain.

The thickness of deposits will decrease away from the

channel as floods deposit most of their load near the

channel banks, and through time it will build up a very

low amplitude lobate body on the basin floor. Areas

adjacent to the fluvial sediment lobe will be at a lower

elevation, and consequently avulsion of the active channel


7/28/2019 1-s2.0-S0037073806002077-main

10/16

will occur, possibly at any point along the river from the

apex down to the terminal splay. Repetition of this process

will result in the active channel occupying different

positions on the alluvial plain through time, with each

avulsion redirecting the river to a nearby lower elevation

course. This will eventually lead to the formation of a fanof fluvial sediments, a process recognised as the

mechanism for the formation of megafans (see below,

Wells and Dorr, 1987; Gohain and Parkash, 1990). At any

one time, most of the system will be an inactive area, and

these abandoned sectors will be subject to pedogenic

modification and reworking of the surface by local run-off

or aeolian activity.

3.7. Bifurcation of channels

Previous models of fluvial distributary systems haveindicated that there is bifurcation of channels down-

stream (Friend, 1978; Nichols, 1987, 1989; Kelly and

Olsen, 1993). Channel bifurcation is present in the distal

zones of modern rivers which have terminal splays such

as the Neales River (Lang et al., 2004) and the

Markanda Fan (Parkash et al., 1983). Bifurcation of

active river channels on a larger scale occurs in the

Okavango Fan (Stanistreet and McCarthy, 1993), which

has an almost horizontal depositional profile, but is not

reported from the steeper active tracts of modern rivers.

It is not possible to establish if two palaeochannels were

coevally flowing, so it is not known if the fluvialdistributary systems considered here would have had

multiple active channels over large distances, but

comparison with modern rivers suggests that it is

unlikely, and more probable that bifurcation only

occurred in the distal zones.

3.8. Proximal to distal extent of channels

The distance that a river will flow out into a basin in an

arid setting will be determined by the balance between the

rate of water supply from the drainage area and the lossesalong the way due to soak-away and evaporation.

Sediment transport will be limited by these same factors

and hence the proximal to distal extent of a distributary

system will be determined by the climate in the hinterland,

which governs the water supply, and the climate in the

basin, which affects the rate of evaporation and the nature

of the floodplain sediment, which will affect rates of

infiltration. Variations in climate over both long and short

time scales will result in a range of distances between the

system apex and the terminal splays as the deposit builds

up, and an interfingering of the proximal and medial, and

medial and distal, zones is to be expected. Awhole system

may prograde or retrograde as the balance between supply

and losses varies: this has been documented in the Luna

System, Spain (Arenas et al., 2001).

4. Fluvial distributary systems and lakes

The conceptual model for fluvial distributary systems

presented in Figs. 1 and 2 is valid for conditions where

the river channels do not reach a basin centre lake.

During periods of high discharge, unconfined flow from

the terminal splays may spread out onto the alluvial plain

and deposit suspended sediment out of a temporary

standing body of water (Fisher et al., 2006-this volume).

This body of water may be considered to be transitional

to an ephemeral lake, depending on where the distinction

between a temporary pond of water on an alluvial plain

as a result of flooding and an ephemeral lake is drawn. Ifthere are changes in the climate which either increases

the water supply from the hinterland or reduces the

evaporation in the basin (or both), then the alluvial plain

will become a lake. This scenario is envisaged in the

Devonian Clair Basin, in which transitions from

fluvially-dominated units to lacustrine units within the

vertical succession are attributed to climatic controls

which determined the presence of, and extent of, a basin-

centre lake (Nichols, 2005). Periods of higher lake level

are also recognised in the Miocene of the Ebro Basin

(Arenas et al., 2001) resulting in an interfingering of

lacustrine and distal fluvial facies.Once a lake is established in the basin the river

system will no longer be terminal, and in place of

terminal splays the channels will feed into a low relief

delta, a lacustrine floodplain delta (Blair and McPher-

son, 1994) such as the present day Volga Delta in the

Caspian Sea (Kroonenberg et al., 1997; Overeem et al.,

2003). Variations in lake level resulting in alternations in

fluvial and lacustrine facies in a similar deltaic setting in

the Pliocene of Azerbaijan have been recognised by

Hinds et al. (2004). With lake level rise, the distal zone

of a fluvial distributary system will be flooded andreplaced by lake margin facies (Fisher et al., 2006-this

volume), but the processes operating in the channels of

medial and proximal zones will be largely unaffected.

The raising of the water table will exert some influence

of soil formation in the overbank areas, but otherwise

the fluvial processes in the tracts of the river which are

not influenced by the lake level will be the same whether

the lake is present or not. A drying of the climate, a

retreat of the lake shoreline and a subsequent fall in lake

level will result in a return to the terminal character of

the fluvial system, a state which is recognised in the

Neales River in central Australia, where terminal splays


7/28/2019 1-s2.0-S0037073806002077-main

11/16

are intermittently active on the former lake bed (Lang et

al., 2004). This potential for fluctuation between a

terminal fluvial and a lacustrine setting for the outer

fringe of a fluvial distributary system is an issue which

has to be taken into account when attempting to decide

on appropriate terminology for these systems (seebelow).

5. Conditions for the formation of a fluvial

distributary system: tectonic and climatic setting

Basins of internal drainage can form in a variety of

tectonic settings, from extensional rift basins, such as

the modern basins in the East African Rift Valley, to

transtensional settings, such as the modern Dead Sea

and the Devonian Orcadian Basin in Scotland, to

foreland basins such as the Ebro Basin, Spain in themid-Cenozoic. Accumulation of a thick succession of

fluvial deposits would be facilitated by subsidence and

a high basin margin sill, which prevents the

development of an external drainage (Fig. 5). A

number of the examples of fluvial distributary systems

in the literature are from the Old Red Sandstone

(ORS) of what is now the North Atlantic area (Fig.

3a): these ORS deposits now outcrop in East Green-

land, West Norway, Scotland and Ireland (Friend et

al., 2000). The basins formed in the post-Caledonian

phase, and lay in the middle of a large continental area

created by the closure of the Iapetus Ocean. They areextensional and transtensional basins within and

adjacent to the Caldeonian orogenic belt which

provided the source of detritus to fill the basins with

thousands of metres of clastic sediment.

During the Devonian the region where these basins

formed lay close to the equator and in an intraconti-

nental setting (Friend et al., 2000): the palaeoclimate

would have been relatively hot, but the nearby

mountains are likely to have been cooler areas with

higher precipitation. Fluvial distributary systems canonly form in basins which have a relatively warm and

dry climate, where water losses through evapo-transpi-

ration exceed the input from direct precipitation and

rivers entering the basin. Desert basins are therefore the

most likely settings, but only if they are adjacent to an

upland area which has a much higher precipitation. A

moderate supply of water is required to establish and

maintain the rivers of the distributary system, and if this

is not present, waterlain deposits are likely to be

restricted to smaller alluvial fans at the basin margin

where deposition is by debris flow or sheetfloodprocesses (e.g. Death Valley, California). On the other

hand, if an excess of water is supplied to an endorheic

basin a lake will form and the river system will feed a

lake delta. Favourable settings are therefore basins

adjacent to, but in the rain-shadow of, mountainous

areas. If the Ebro Basin was still endorheic, the modern

climatic and tectonic setting may be suitable: rainfall in

the Pyrenees is between 1000 and 2000 mm.a1, but the

centre of the basin is semi-arid, with precipitation of

around 300 mm.a1 and summer temperatures reaching

over 40 C.

The supply of sediment to the distal area must exceedthe rate basin-floor subsidence. This will ensure that the

base-level in the basin rises as a result of basin-filling

with sediment, and, consequently, the river gradient will

be very low, tending to horizontal. Low gradient will

Fig. 5. Tectonic and climatic setting for the formation of fluvial distributary systems.


7/28/2019 1-s2.0-S0037073806002077-main

12/16

tend to promote avulsion because deposition around the

active channel will create a lobe on the alluvial plain and

the river may follow a new, steeper gradient path to the

side of the lobe and thus create the fan shape that many

fluvial distributary systems possess. An aggradational

pattern of strata will form as sedimentary fill of the basinresults in a rise in the base-level, provided that the

discharge remains approximately the same and the

fluvial system has the same extent into the basin.

6. Fluvial distributary systems, terminal fans,

fluvial fans, megafans and lake deltas

6.1. Alluvial fans, fluvial fans, megafans, humid fans

The problem of terminology to be used in the

description of fan-shaped bodies of alluvium has exer-cised a number of geologists and geomorphologists over

the years. Blair and McPherson (1994) consider that the

term alluvial fan should be restricted to steep, debris

flow and sheetflood dominated deposits, and draw a

distinction between the processes and products of these

alluvial fans and those of rivers. Such a distinction is not

recognised by all workers in this field, and there are

many examples of alluvial fans in the literature which

are the products of deposition by rivers (e.g. Nemec and

Postma, 1993). Some authors refer to fans formed by the

migration of channelised streams as fluvial fans:

Collinson (1996) uses this term to encompass depositsranging from small, steep fans as described by Nemec

and Postma (1993) to the megafans of the Ganges

valley, such as the Kosi Fan documented by Wells and

Dorr (1987) and Gohain and Parkash (1990) and the

Ganga Megafan (Shukla et al., 2001). The term fluvial

fan or fluvial megafan has been used for fluvial

distributary systems, for example Arenas et al. (2001)

for the Luna System in the Ebro Basin, Pennsylvanian

deposits in the Paradox Basin, USA (Barbeau, 2003)

and modern and Tertiary deposits in the central Andes

(Horton and DeCelles, 2001). The modern megafansare characterised by channels that have a through-flow

of water and discharge is continuous into a channel

downstream of the area of the fan: the fan morphology is

a result of a change in slope promoting avulsion in a

particular tract of the river. Use of the term humid fan

was originally coined in the context of fans of glacial

outwash braidplain deposits (Boothroyd and Numme-

dal, 1978): recent work on the relationship between

climate and processes on alluvial fans has shown that the

concept of wet and dry fans is oversimplistic, and

sometimes misleading (Harvey and Wells, 1994; Harvey

et al., 2005).

6.2. Subaerial and losimean fans

Perhaps the most encompassing term is subaerial fan,

which Stanistreet and McCarthy (1993) use to include

debris flow alluvial fans, braided river fans like the Kosi

Fan, and what they term losimean fans, of which thepresent day Okavango Fan in Botswana is their type

example. The Okavango Fan is a large (150 km radius),

very low gradient system with multiple distributary

channels, some of which are active, others in the process

of abandonment (Stanistreet and McCarthy, 1993). In the

proximal reaches the rivers are sandy and meandering,

becoming dominated by more confined single and

anastomosing channels in the medial zone before flowing

to a distal zone which is poorly channelised and dominated

by unconfined flow. Discharge decreases downstream due

to evaporation (and transpiration from lush vegetation inthe interchannel areas) and only 2% of the input flow exits

from the edge of the fan to flow in a single channel to a

lake. The Okavango Fan is therefore an almost terminal

system, and, as Stanistreet and McCarthy (1993) observe,

there are a number of similarities with the Luna System as

documented byNichols (1987). Use of the term losimean

fan is limited by its definition as a fan dominated by

meandering and low sinuosity rivers with a very low

depositional gradient (Stanistreet and McCarthy, 1993).

6.3. Terminal fans

Kelly and Olsen (1993) use the term terminal fan for

fluvial distributary systems which have no surface flow

into a lake or the sea. A modern case study comes from the

Markanda River in India, which is an ephemeral river that

ends in a series of distributary channels which form a

terminal fan (Parkash et al., 1983). The riverloses most of

its discharge before it reaches the terminal fan where it

bifurcates into a series of distributary channels and

subfans. There is evidence of frequent avulsion building

up a lobate sheet of sandy channel and muddy overbank

deposits over an area of 64 km2

, with an almost continuouslayer of channel facies preserved in the top 2.5 m of the fan

(Parkash et al., 1983). The Markanda Fan case study was

one of only two modern examples used in Kelly and

Olsen's (1993) review, the other being the Gash River Fan

in Sudan (Abdullatif, 1989). As Kelly and Olsen (1993)

point out, the only fan for which there is any detailed

sedimentological data, the Markanda fan, is considerably

smaller than most of the ancient examples they selected.

A further consideration in trying to find appropriate

terminology is that the characteristics of the distal parts of

the system will vary depending on the level of water in the

basin centre. Under conditions where evaporative losses


7/28/2019 1-s2.0-S0037073806002077-main

13/16

exceed the supply of water through direct precipitation and

via the river systems, the basin centre will be a dry

evaporative alluvial plain, and the channels will end in

terminal splays. However, during a wetter climatic phase

the lake level will rise and the distal ends of the river

channels will flow directly into the water body to form alake delta. In this state the fluvial system cannot be

regarded as terminal, like the Markanda Fan ofParkash et

al. (1983) or in the sense envisaged by Kelly and Olsen

(1993). If these lake level fluctuations are relatively small,

only the distal part of the fluvial system will be affected:

the characteristics of channel and overbank sedimentation

in proximal and medial areas will be the same whether the

river is flowing into a lake or ending in a terminal splay.

However, if the terminal fan moniker is to be used for the

whole system, it can strictly only be applied to the times

when the river does not reach the lake, but this distinctioncannot be determined for the proximal and medial parts of

the system.

6.4. Lake deltas

In the classification of river and alluvial fan deposits

suggested by Blair and McPherson (1994) the fluvial

distributary systems described in this paper could be

considered to be ephemeral-lacustrine floodplain del-

tas. The drawback of using this approach is that a

fluvial system prograding into a basin which is a dry

alluvial plain for almost all the time, and ephemerallakes only form away from the distal fringes of the river

channels, does not really conform to the general view of

a delta. The term would be appropriate for wetter

stages of the basin history in examples such as the Luna

and Huesca Systems, but evidence for the wetter climate

and the presence of a lake cannot necessarily be

recognized in the bulk of the fluvial deposits.

6.5. Fluvial distributary systems

The approach preferred here is to use the term fluvialdistributary system to describe the form of a river

system which has the shape of a fluvial fan, has a

downstream decrease in discharge and has a distal area

of eitherterminal splays, when there is no lake present,

or forms a floodplain lake delta at times of high lake

level. If it has a very low gradient it may also be a

losimean fan in the scheme of classification of

subaerial fans suggested by Stanistreet and McCarthy

(1993). A river pattern may be described as distribu-

tary in a variety of settings, for example as part 1of a

delta, but the term fluvial distributary system has only

been applied to the river systems with the characteristics

originally defined by Friend (1978) and further

described by Graham (1983), Hirst and Nichols

(1986), Alonso Zarza et al. (1993) and Mather (2000).

7. The stratigraphic architecture of fluvial

distributary system deposits

The short and long-term preservation potential of

fluvial distributary system deposits in endorheic basins is

good. To preserve thick successions of fluvial strata the

area of deposition must have a long history of relative

base-level rise. This condition is met in basins of internal

drainage which are underfilled (Bohacs et al., 2000) and

have rates of evaporation exceeding water supply.

Accumulation of strata in the basin will continue until

the spill point is reached, and the total thickness will be

greater than the original accommodation up to the spillpoint as a result of subsidence due to sedimentary

loading, and there may be continued tectonic subsidence

in the basin. Many of the Devonian endorheic basins

discussed above underwent syn-depositional subsidence

due to continued tectonic activity (e.g. the Hornelen

Basin in West Norway, Nilsen and McLaughlin, 1985).

There is therefore the potential to accumulate many

thousands of metres of fluvial deposits in these settings

and many of the examples from the stratigraphic record

are successions of considerable thickness.

On the basis of the conceptual model (Figs. 1 and 2)itis

possible to predict the stratigraphic architecture of thicksuccessions of fluvial distributary system deposits. In some

respects, fluvial distributary systems show similarities with

more conventional rivers of similar scales. There is an

overall decrease in grain size downstream as the rivers

undergo a transition from bedload-dominated to a mixed-

load system. There is also an increase in the proportion of

overbank deposits preserved. However, there are also

elements of fluvial distributary systems which make them

significantly different from normal river deposits and

these should be taken into account when constructing a

palaeogeographic scenario or developing a predictiveframework of reservoir characteristics.

The overall shape of the sediment body will be fan-

shaped if the basin morphology allows the channels to

avulse in a radial pattern across the alluvial plain.

However, the degree to which a system spreads out will

be governed by the basin geometry: a river feeding into

one end of an elongate basin will form a narrower fan

because the lateral avulsion of the channels will be

limited by the sides of the basin. The proximal to distal

dimensions will be largely determined by the climate in

both the hinterland and the basin controlling the

discharge in the rivers and the rates of evaporation.


7/28/2019 1-s2.0-S0037073806002077-main

14/16

The case studies from the Ebro Basin show that as

well as the depth of the channels decreasing

downstream the width of the sandstone bodies formed

as channel deposits also decreases. In addition there is

also a distal decrease in the percentage of in-channel

component of the deposits, partly because of thesmaller size of the channels, but also this is to be

expected in these examples because the radial pattern

spreads the channel deposits over a wider area distally.

The consequence of these trends is that the degree of

interconnectedness of channel sandstone bodies

decreases in an exponential way from the proximal

to distal areas in the Luna and Huesca Systems. The

decrease in in-channel component downstream will be

less extreme in systems with a less radial pattern, but

nevertheless this is a characteristic feature which is

important to reservoir modelling.In simple models of fluvial sedimentation, coarse

sediment is mainly deposited as bedload in the

channels, and finer material from suspension on the

floodplain. However, in the distal parts of fluvial

distributary systems, the partition of sedimentation is

not so straightforward. Evidence from the Ebro Basin

shows that some of the channel scours are wholly or

partly filled with mudrock (Fig. 4c, d), and that the

floodplain facies association includes thin sheets of

sandstone, which may locally make up a quarter of the

succession. Therefore, in the distal area, more sand is

to be found as sheet floodplain facies than as channel-fill. Vertical connection between sheets separated by

mudrocks only occurs where they are incised by a

sand-filled channel, and these are relatively uncom-

mon. The prospects for a reservoir of interconnected

sandstone sheets are therefore poor in the distal parts

of a system.

Rivers which are hydrologically connected to the

ocean are influenced by changes in relative sea level: a

relative sea level fall results in valley incision and lateral

confinement of the fluvial tract (Shanley and McCabe,

1994). In endorheic basins relative sea level has noeffect on the rivers and in a relatively arid climate any

basin-centre lake is likely to be relatively shallow. As

noted above, fluctuations in the lake level of only a few

metres may have little effect on the proximal and medial

parts of the fluvial tract. Overall the base-level in an

endorheic basin will rise through time, if there is no

differential uplift and subsidence at the basin margin

(Nichols, 2004), so the pattern of sedimentation will be

aggradational. A further characteristic of a fluvial

distributary system in an endorheic basin is therefore

an absence of incised valley fill, a feature originally

recognised by Friend (1978).

8. Conclusions

Fluvial distributary systems are a distinctive style of

continental sedimentation that occur mainly in basins of

internal drainage. They are characterised by a down-

flow decrease in channel dimensions caused by a loss ofdischarge downstream due to evaporation and soak-

away. They are typically fan-shaped bodies tens of

kilometres in radius built up by repeated avulsion of

river channels on a very low gradient alluvial plain.

Terminal splays are considered to be elements of

these systems, being the areas of unchannelised flow

and deposition at the distal ends of active channels

which terminate on the alluvial plain. At times of lake

highstand the distal zone of the system may be

indundated and the river channels will feed a lacustrine

delta. These entire fluvial systems cannot necessarily becalled terminal fans because at times of lake highstand

the systems are not terminal. They can be considered to

be a type of fluvial fan or megafan, but differ from

the type examples of megafans which have through-

flowing river systems beyond the limits of the fan body.

Most have the plan form of a subaerial fan, but are not

necessarily fan-shaped bodies if restricted by basin

topography. The term fluvial distributary system is

therefore preferred on the grounds of previous usage and

avoiding the need to redefine any terminology.

Potential reservoir properties will vary considerably

across the system: in the proximal zone coarse channelfacies dominate and are likely to be fully interconnected;

across the medial zone the size and proportion of channel

sandstone decrease, but some connectivity may be

provided by overbank sheet sandstones; distal areas are

characterised by low concentrations of smaller channels

and much of the sand deposited as unconnected sheets.

Acknowledgements

GJN would like to acknowledge the contribution of

colleagues who have contributed to discussions aboutthese systems, particularly Peter Friend, Philip Hirst,

Colin North and Ed Williams. JAF acknowledges the

support of an NERC studentship. Kevin Bohacs and

Anne Mather are thanked for their constructive reviews.

References

Abdullatif, O.M., 1989. Channel-fill and sheet-flood facies sequences

in the ephemeral terminal River Gash, Kassala, Sudan. Sedimen-

tary Geology 63, 171184.

Alonso Zarza, A.M., Calvo, J.P., Garca del Cura, M.A., 1993.

Palaogeomorphological controls on the distribution and sedimen-tary styles of alluvial systems, Neogene of the NE of the Madrid


7/28/2019 1-s2.0-S0037073806002077-main

15/16

Basin, (central Spain). In: Marzo, M., Puigdefbregas, C. (Eds.),

Alluvial Sedimentation, Special Publication of the International

Association of Sedimentologists, vol. 17, pp. 277292.

Arenas, C., Milln, H., Pardo, G., Pocov, A., 2001. Ebro Basin

continental sedimentation associated with late compressional

Pyrenean tectonics (north-eastern Iberia): controls on basin margin

fans and fluvial systems. Basin Research 13, 6589.Barbeau, D.L., 2003. A flexural model for the Paradox Basin:

implications for the tectonics of the Ancestral Rocky Mountains.

Basin Research 15, 97115.

Blair, T.C., McPherson, J.G., 1994. Alluvial fans and their natural

distinction from rivers based on morphology, hydraulic processes,

sedimentary processes and facies assemblages. Journal of

Sedimentary Research. Section A, Sedimentary Petrology and

Processes 64, 450489.

Bohacs, K.M., Carroll, A.R., Neal, J.E., Mankiewicz, P.J., 2000. Lake

basin type, source potential and hydrocarbon character: an

integrated sequence-stratigraphic-geochemical framework. In:

Gierlowski-Kordesch, E.H., Kelts, K.R. (Eds.), Lake Basins

Through Space and Time. American Association of Petroleum

Geologists Studies in Geology, vol. 46, pp. 334.Boothroyd, J.C., Ashley, G.M., 1975. Processes, bar morphology and

sedimentary structures in braided outwash fans, north eastern Gulf

of Alaska. In: Jopling, A.V., McDonald, B.C. (Eds.), Glaciofluvial

and Glaciolacustrine Sedimentation. Society for Economic

Paleontologists and Mineralogists Special Publication, vol. 23,

pp. 193222.

Boothroyd, J.C., Nummedal, D., 1978. Proglacial braided outwash: a

model for humid alluvial-fan deposits. In: Miall, A.D. (Ed.),

Fluvial Sedimentology. Canadian Society of Petroleum Geologists

Memoir, vol. 5, pp. 641668.

Collinson, J.D., 1996. Alluvial sediments, In: Reading, H.G. (Ed.),

Sedimentary Environments: Processes, Facies, Stratigraphy, 3rd

ed. Blackwell Science, Oxford, pp. 3782.

Fisher, J.A., Nichols, G.J., Waltham, D.A., 2006-this volume.

Unconfined flow deposits in distal sectors of fluvial distributary

systems: examples from the Miocene Luna and Huesca

Systems, northern Spain Sedimentary Geology. doi:10.1016/j.

sedgeo.2006.07.005.

Fraticelli, C.M., West, B.P., Bohacs, K.M., Patterson, P.E., Heins,

W.A., 2004. Vegetation-precipitation interactions drive palaeoen-

vironmental evolution. Eos Transactions of the American

Geophysical Union 85 (47) (Fall Meet. Suppl., Abstract

H54A-06).

Friend, P.F., 1978. Distinctive features of some ancient river systems.

In: Miall, A.D. (Ed.), Fluvial Sedimentology. Canadian Society of

Petroleum Geologists Memoir, vol. 5, pp. 531542.

Friend, P.F., 1989. Space and time analysis of river systems, illustratedby Miocene systems of the northern Ebro Basin in Aragon, Spain.

Revista de la Sociedad Geolgica de Espaa 2, 5564.

Friend, P.F., Moody-Stuart, M., 1972. Sedimentation of the Wood Bay

Formation (Devonian) of Spitsbergen: regional analysis of a late

orogenic basin. Norsk Polarinstitutt Skrifter 157, 177.

Friend, P.F., Slater, M.J., Williams, R.C., 1979. Vertical and lateral

building of river sandstone bodies, Ebro Basin, Spain. Journal of

the Geological Society (London) 136, 3946.

Friend, P.F., Alexander-Marrack, P.D., Allen, K.C., Nicholson, J.,

Yeats, A.K., 1983. Devonian sediments of East Greenland VI:

review of results. Meddelelser om Grnland 206 (6) (96 pp.).

Friend, P.F., Williams, B.P.J., Ford, M., Williams, E.A., 2000.

Kinematics and dynamics of Old Red Sandstone basins. In:

Friend, P.F., Williams, B.P.J. (Eds.), New Perspectives on the Old

Red Sandstone. Special Publications, vol. 180. Geological Society,

London, pp. 2960.

Gohain, K., Parkash, B., 1990. Morphology of the Kosi Megafan. In:

Rachoki, A.H., Church, M. (Eds.), Alluvial Fans: A Field

Approach. John Wiley, Chichester, pp. 151178.

Graham, J.R., 1983. Analysis of the Upper Devonian Munster Basin,

an example of a fluvial distributary system. In: Collinson, J.D.,Lewin, J. (Eds.), Modern and Ancient Fluvial Systems, Special

Publication, International Association of Sedimentologists, vol. 6,

pp. 473484.

Harvey, A.M., Wells, S.G., 1994. Late Pleistocene and Holocene

changes in hillslope sediment supply to alluvial fan systems:

Zzyzx, California. In: Millington, A.C., Pye, K. (Eds.), Environ-

mental Change in Drylands: Biogeographical and Geomorpholog-

ical Perspectives. Wiley, Chichester, pp. 6784.

Harvey, A.M., Mather, A.E., Stokes, M., 2005. Alluvial fans:

Geomorphology, Sedimentology, dynamics-Introduction: a review

of alluvial fan research. In: Harvey, A.M., Mather, A.E., Stokes,

M. (Eds.), Alluvial Fans: geomorphology, sedimentology, dyna-

micsGeological Society, London, Special Publications, vol. 251,

pp. 17.Hinds, D.J., Aliyeva, E., Allen, M.B., Davies, C.E.,Kroonenberg, S.B.,

Simmons, M.D., Vincent, S.J., 2004. Sedimentation in a discharge

dominated fluvial-lacustrine system: the Neogene Productive

Series of the South Caspian Basin, Azerbaijan. Marine and

Petroleum Geology 21, 613638.

Hirst, J.P.P., 1991. Variations in alluvial architecture across the Oligo-

Miocene Huesca Fluvial System, Ebro Basin, Spain. In: Miall, A.D.,

Tyler, N. (Eds.), The Three-Dimensional Facies Architecture of

Terrigenous Clastic Sediments and its Implications for Hydrocarbon

Discovery and Recovery. SEPM (Society for Sedimentary Geology)

Concepts in Sedimentology and Palaeontology, vol. 3, pp. 111121.

Hirst, J.P.P., Nichols, G.J., 1986. Thrust tectonic controls on alluvial

sedimentation patterns, southern Pyrenees. In: Allen, P.A., Home-

wood, P. (Eds.), Foreland Basins, International Association of

Sedimentologists Special Publication, vol. 8, pp. 153164.

Horton, B.K., DeCelles, P.G., 2001. Modern and ancient fluvial

megafans in the foreland basin system of the central Andes,

southern Bolivia: implications for drainage network evolution in

fold-thrust belts. Basin Research 13, 4363.

Jupp, P.E., Spurr, B., Nichols, G.J., Hirst, J.P.P., 1987. Statistical

estimation of the apex of a sediment distribution system from

palaeocurrent data. Mathematical Geology 19, 319333.

Kelly, S.B., Olsen, H., 1993. Terminal fans a review with reference

to Devonian examples. Sedimentary Geology 85, 339374.

Kroonenberg, S.B., Rusakov, G.V., Svitoch, A.A., 1997. The

wandering of the Volga delta: a response to rapid Caspian sea-

level change. Sedimentary Geology 107, 189209.Lang, S.C., Payenberg, T.H.D., Reilly, M.R.W., Hicks, T., Benson, J.,

Kassan, J., 2004. Modern analogues for dryland sandy fluvial-

lacustrine deltas and terminal splay reservoirs. APPEA Journal

329356.

MacCarthy, I.A.J., 1990. Alluvial sedimentation patterns in the

Munster Basin, Ireland. Sedimentology 37, 685712.

Mather, A.E., 2000. Impact of headwater river capture on alluvial

system development. Journal of the Geological Society (London)

157, 957966.

Nemec, W., Postma, G., 1993. Quaternary alluvial fans in southwest-

ern Crete: sedimentation processes and geomorphic evolution. In:

Marzo, M., Puigdefbregas, C. (Eds.), Alluvial Sedimentation-

Special Publications, International Association of Sedimentolo-

gists, vol. 17, pp. 235276.

http://dx.doi.org/10.1016/j.sedgeo.2006.07.005http://dx.doi.org/10.1016/j.sedgeo.2006.07.005http://dx.doi.org/10.1016/j.sedgeo.2006.07.005

7/28/2019 1-s2.0-S0037073806002077-main

16/16

Nichols, G.J., 1987. Structural controls on fluvial distributary

systemsthe Luna System, Northern Spain. In: Ethridge, F.G.,

Florez, R.M., Harvey, M.D. (Eds.), Recent Developments in

Fluvial Sedimentology. Special Publication, vol. 39. Society of

Economic Palaeontologists and Mineralogists, pp. 269277.

Nichols, G.J., 1989. Structural and sedimentological evolution of part

of the west central Spanish Pyrenees in the Late Tertiary. Journal ofthe Geological Society (London) 146, 851857.

Nichols, G.J., 2004. Sedimentation and base level controls in an

endorheic basin: the Tertiary of the Ebro Basin, Spain. Boletn

Geolgico y Minero, Espaa 115, 427438.

Nichols, G.J., 2005. Sedimentary evolution of the Lower Clair Group,

Devonian, west of Shetland: climate and sediment supply controls

on fluvial, aeolian and lacustrine deposition. In: Dor, A.G.,

Vining, B.A. (Eds.), Petroleum Geology: North West Europe and

Global Perspectives - Proceedings of the 6th Petroleum Geology

Conference. Geological Society, London, pp. 957967.

Nilsen, T.H., McLaughlin, R.J., 1985. Comparison of tectonic

framework and depositional patterns of the Hornelen strikeslip

basin of Norway and the Ridge and Little Sulphur Creek strike

slip basins of California. In: Biddle, K.T., Christie-Blick, N. (Eds.),Strike Slip Deformation, Basin Formation and Sedimentation.

Society of Economic Palaeontologists and Mineralogists Special

Publications, vol. 37, pp. 79103.

Overeem, L., Kroonenberg, S.B., Groenesteijn, A., Svitoch, G.V., 2003.

Small-scale stratigraphy in a large ramp delta: recent and Holocene

sedimentation in the Volga delta, Caspian Sea.Sedimentary Geology

159, 133157.

Parkash, B., Awasthi, A.K., Gohain, K., 1983. Lithofacies of the

Markanda terminal fan, Kurukshetra district, Haryana, India. In:

Collinson, J.D., Lewin, J. (Eds.), Modern and Ancient Fluvial

Systems. International Association of Sedimentologists Special

Publication, vol. 6, pp. 337344.

Richmond, L.K., Williams, B.P.J., 2000. A new terrane in the Old Red

Sandstone of the Dingle Peninsula, SW Ireland. In: Friend, P.F.,

Williams, B.P.J. (Eds.), New Perspectives on the Old Red Sand-

stone. Special Publications, vol. 180. Geological Society, London,

pp. 147184.

Sadler, S.P., Kelly, S.B., 1993. Fluvial processes and cyclicity in

terminal fan deposits: an example from the Late Devonian of

southwest Ireland. Sedimentary Geology 85, 375386.

Shanley, K.W., McCabe, P.J., 1994. Perspectives on the sequencestratigraphy of continental strata. American Association of

Petroleum Geologists Bulletin 78, 544568.

Shukla, U.K., Singh, I.B., Sharma, M., Sharma, S., 2001. A model of

alluvial megafan sedimentation: Ganga Megafan. Sedimentary

Geology 144, 243262.

Stanistreet, I.G., McCarthy, T.S., 1993. The Okavango Fan and the

classification of subaerial fan systems. Sedimentary Geology 85,

115133.

Tooth, S., 1999a. Floodouts in Central Australia. In: Miller, A.J.,

Gupta, A. (Eds.), Varieties of Fluvial Form. Wiley and Sons,

London, pp. 219247.

Tooth, S., 1999b. Downstream changes in floodplain character on the

Northern Plains of arid central Australia. In: Smith, N.D., Rogers, J.

(Eds.), Fluvial Sedimentology VI. Special Publication of theInternational Association of Sedimentologists, vol. 28, pp. 93112.

Wells, N.A., Dorr Jr., J.A., 1987. A reconnaissance of sedimentation

on the Kosi alluvial fan of India. In: Ethridge, F.G., Florez, R.M.,

Harvey, M.D. (Eds.), Recent Developments in Fluvial Sedimen-

tology, Society of Economic Palaeontologists and Mineralogists

Special Publication, vol. 39, pp. 5161.

Williams, E.A., 2000. Flexural cantilever models of extensional

subsidencein the Munster Basin (SE Ireland) and Old Red Sandstone

fluvial dispersal systems. In: Friend, P.F., Williams, B.P.J. (Eds.),

New perspectives on the Old Red Sandstone. Special Publications,

vol. 180. Geological Society, London, pp. 239268.


Documents

1-s2.0-S0037073806002077-main